Recent data suggest that the adaptive immune system is basically self-referential. This is well established for T cells, which recognize antigen only when it is presented by the MHC molecules that these T cells have been selected upon. It is far less clear what role self-recognition plays in the onward survival of T cells or in their activation. Even more obscure is the positive selection of mature B cells we have recently reported. We propose to explore these issues in four specific aims in this grant. These are as follows: 1. What is the role of self- peptide: self-MHC complexes in the development of the mature T cell receptor repertoire in the thymus? Do the same rules apply for MHC class I molecules as were previously observed with MHC class II molecules? We hypothesize that self-recognition applies to all MHC class II-restricted TCRs, and we will confirm this by testing several examples, as well as 3 MHC class I-restricted TCRs to see if MHC class I restricted TCRs obey the same rules. 2. What is the role of self-peptide: self-MHC complexes in the periphery in maintaining the mature, peripheral T cell receptor repertoire? We hypothesize that this interaction sustains T cells in the periphery, and that this process is accompanied by slow turn-over of mature T cells. 3. Does self-peptide: self-MHC recognition contribute to antigen stimulation? We hypothesize that the ability of small numbers of foreign peptides to trigger T cells is due to their ability to cause tight adhesion between a T cell and an antigen-presenting cell, leading to effective engagement of the T cell receptor with self-peptide: self-MHC complexes. 4. What is responsible for the positive selection of B cells, and for the stable maintenance of mature B cells in the periphery? We hypothesize that it is self-antigen recognition at a sub-threshold level, leading to a highly restricted set of B cell receptors. We will test these hypotheses in a variety of ways to determine their correctness.